Identifying Sequence\nEffects on Chain Dimensions of\nDisordered Proteins by Integrating Experiments and Simulations

It has become increasingly evident that the conformational\ndistributions\nof intrinsically disordered proteins or regions are strongly dependent\non their amino acid compositions and sequence. To facilitate a systematic\ninvestigation of these sequence-ensemble relationships, we selected\na set of 16 naturally occurring intrinsically disordered regions of\nidentical length but with large differences in amino acid composition,\nhydrophobicity, and charge patterning. We probed their conformational\nensembles with single-molecule Förster resonance energy transfer\n(FRET), complemented by circular dichroism (CD) and nuclear magnetic\nresonance (NMR) spectroscopy as well as small-angle X-ray scattering\n(SAXS). The set of disordered proteins shows a strong dependence of\nthe chain dimensions on sequence composition, with chain volumes differing\nby up to a factor of 6. The residue-specific intrachain interaction\nnetworks that underlie these pronounced differences were identified\nusing atomistic simulations combined with ensemble reweighting, revealing\nthe important role of charged, aromatic, and polar residues. To advance\na transferable description of disordered protein regions, we further\nemployed the experimental data to parametrize a coarse-grained model\nfor disordered proteins that includes an explicit representation of\nthe FRET fluorophores and successfully describes experiments with\ndifferent dye pairs. Our findings demonstrate the value of integrating\nexperiments and simulations for advancing our quantitative understanding\nof the sequence features that determine the conformational ensembles\nof intrinsically disordered proteins.